REPETITIVE STIMULATION-INDUCED POTENTIATION OF EXCITATORY TRANSMISSION IN THE RAT DORSAL HORN - AN IN-VITRO STUDY

1. The effects of repetitive stimulation of primary afferents in lumbar dorsal roots on synaptic transmission in the dorsal horn (DH) were studied in a rat spinal cord slice-dorsal root ganglion (DRG)-peripheral nerve trunk preparation by the use of intracellular recording from neurons (n = 115) of the spinal dorsal horn (depth 147 +/- 139, mean +/- SD). All DH neurons were excited synaptically by electrical stimulation of the dorsal root or the peripheral nerve trunk. The electrical shocks were calibrated to produce activation either of large fibers (10-20 V, 0.02 ms) or the whole fiber population including unmyelinated afferents (supramaximal stimulus: >35 V, 0.5 ms). Postsynaptic potentials induced by low intensity repetitive stimulation of primary afferents at frequencies below 5 Hz failed to produce a prolonged change in the resting membrane potential. In 97/115 DH neurons, slow excitatory postsynaptic potentials (EPSP)-evoked by high intensity low-frequency repetitive stimulation (0.1-2 Hz) of primary afferents-summated, producing a prolonged cumulative depolarization. In the remaining 18/115 DH neurons, high intensity low-frequency stimulation produced a cumulative hyperpolarizing response. 2. In 22 of 97 neurons that responded to high intensity repetitive stimulation with a cumulative depolarization, windup in the firing of action potentials was recorded. In all but two experiments, neurons that responded with windup to stimulation of one root responded with windup to stimulation of the adjacent dorsal root. In 14/22 windup neurons, the synaptic response to high intensity stimulation of primary afferents was composed of a short latency EPSP, followed by an inhibitory postsynaptic potential (IPSP), followed by a slow EPSP. The decrease of the amplitude and duration of the IPSP obtained during train stimulation did not seem to contribute to facilitation of transmission induced by repetitive stimulation. 3. The windup in firing of action potentials was followed by a prolonged potentiation of synaptic transmission in tetanized synapses. A test of other, adjacent primary afferents revealed that these synapses in the neurons in the superficial laminae had not undergone potentiation. This ''synaptic specificity'' of postwindup potentiation suggested that the mechanism for the induction of stimulation-dependent changes in the excitability of the DH neuron is presynaptic to the recorded-from neuron. 4. In a concentration of 0.5 mu M and higher, tetrodotoxin (TTX) applied to sensory neurons selectively blocked action potentials in large myelinated primary afferents. In addition to completely eliminating the lower threshold fiber-evoked fast EPSPs in all of the neurons tested (n = 12), TTX depressed, but in the majority of neurons failed to block, repetitive stimulation-induced cumulative depolarization and windup in the action potential discharge by DH neurons. 5. Superfusion of 1-10 mu M capsaicin onto DRG neurons resulted in a long-lasting excitation of the DH neurons. Most of the DH neurons, subsequent to the application of 10 mu M capsaicin to the sensory neurons, exhibited a long-lasting desensitization (n = 11). In addition to the lack of responsiveness to repetitive applications of capsaicin, the slow component of the composite EPSP evoked by activation of whole primary afferents was blocked, and repetitive stimulation-evoked prolonged depolarization and windup were eliminated. 6. These findings confirmed that activation of high threshold afferents is necessary for the use-dependent alterations in the synaptic response in the spinal cord manifested as facilitation and windup. The windup in firing of action potentials was indirectly dependent on the frequency of repetitive stimulation through depending on the amplitude of cumulative membrane depolarization. The ''synaptic specificity'' of postwindup potentiation suggests a mechanism for the induction of stimulation-dependent changes in the excitability of the DH neuron that is presynaptic to the recorded windup neuron.